Pancreatic cancer has a dismal prognosis due in large part to the ineffectiveness of current treatment regimens. Our long-term goal is to identify key factors underlying pancreatic cancer development so that they can be targeted for new therapeutic strategies. Our hypothesis is that activation of RalA by both K-Ras and Aurora-A kinase is a key step in pancreatic tumorigenesis. This hypothesis is based on the following: 1) Nearly all pancreatic cancers have an activating mutation in the oncogene K-Ras critical for tumorigenesis, and this oncogenic signal is mediated in large part by the RalGEF pathway in humans; 2) the transforming RalGEF effector RalA is crucial for both transformed and tumorigenic growth of pancreatic cancer cell lines; 3) RalA serine 194, which is phosphorylated by Aurora-A kinase, is required for transformation of RalGEF- dependent human cancer cells, and 4) Aurora-A inhibition decreases tumorigenic growth of pancreatic cancer cell lines. We propose to evaluate the roles of RalA and Aurora-A in pancreatic tumorigenesis: Specific Aim 1: Determine if loss of RalA inhibits K-ras-driven pancreatic tumorigenesis in mice. We will evaluate spontaneous pancreatic tumorigenesis in transgenic mice which conditionally express anoncogenic K-ras allele but not RalA in the pancreas. Specific Aim 2: Determine whether RalA functions through its known effectors RalBPI, Exo84, or Sec5 to mediate the development of pancreatic cancer. We will a) perform loss-of-function analysis of the candidate RalA downstream effectors RalBPI, Exo84, or Sec5, and determine the effect on the transformation and tumorigenic growth of pancreatic cancer cell lines, and b) determine whether phosphorylation of RalA S194 alters the binding affinities of RalA for RalBPI, Exo84, or Sec5. Specific Aim 3: Determine whether Aurora-A phosphorylation of RalA S194 enhances pancreaticcancer growth. We will a) establish whether phosphorylation of RalA S194 by Aurora-A is required for transformed and tumorigenic growth in pancreatic cancer cell lines, b) establish whether Aurora-A inhibition requires RalA S194 to decrease growth of pancreatic cancer cells, and c) determine whether RalA S194 phosphorylation causes intracellular relocalization of RalA in pancreatic cancer cells.